基于数据挖掘的火电厂制粉系统能效评估模型研究

Research on Evaluation Model of Energy Efficiency for Pulverizing System based on Data Mining

为降低超超临界机组在运行过程中的能耗,针对机组制粉系统建模过程繁琐及系统能效评估存在人为主观因素强和不确定性等缺陷,提出了一种适用于火电机组制粉系统能效评估的灰色模糊评估模型。通过引入K-medoids聚类算法挖掘出各能效指标的基准值,避免了建模过程繁琐的问题,并有效削弱了现场数据中异常值的影响;通过将灰色模糊评估模型中的关联系数作为隶属度向量,并结合层次分析法(AHP)得到各分层指标的权重系数,有效地避免了评估过程中人为主观因素强和不确定性等问题。计算证明,用该方法能有效对火电机组制粉系统进行能效评估。

To reduce the energy consumption of the ultra-supercritical units during operation,the cumbersome modeling process of coal pulverizing system,and the strong subjective factors and uncertainty in system energy efficiency evaluation,this paper proposes a gray fuzzy evaluation model for energy efficiency evaluation.The K-medoids clustering algorithm is introduced to avoid tedious modeling problems,which also effectively weakens the influence of field outliers.By using the correlation coefficient of grey fuzzy evaluation model as the membership degree vector,the correlation degree of each index is obtained.The membership degree vector is combined with the analytic hierarchy process(AHP) to effectively avoid problems such as strong human subjective factors and uncertainty during the evaluation.The calculation proves that this method can effectively evaluate the energy efficiency of pulverizing system.